ContentslistsavailableatScienceDirect
Industrial
Crops
and
Products
j o u r n al ho me p ag e :w w w . e l s e v i e r . c o m / l o c a t e / i n d c r o p
Estimates
of
repeatability
coefficients
and
selection
gains
in
Jatropha
indicate
that
higher
cumulative
genetic
gains
can
be
obtained
by
relaxing
the
degree
of
certainty
in
predicting
the
best
families
Bruno
Galvêas
Laviola
a,∗,
Ana
Maria
Cruz
e
Oliveira
b,
Leonardo
Lopes
Bhering
b,
Alexandre
Alonso
Alves
a,
Rodrigo
Barros
Rocha
c,
Bruno
Ermelindo
Lopes
Gomes
b,
Cosme
Damião
Cruz
baEmbrapaAgroenergy,ParqueEstac¸ãoBiológica,Brasília70770-901,Brazil bUniversidadeFederaldeVic¸osa,CampusUniversitário,Vic¸osa36570-000,Brazil
cEmbrapaRondônia,BR364,Km5,5,ZonaRural,CaixaPostal127,PortoVelho76815-800,Brazil
a
r
t
i
c
l
e
i
n
f
o
Articlehistory:Received12June2013
Receivedinrevisedform6August2013 Accepted7August2013 Keywords: JatrophacurcasL. Plantbreeding Selection Quantitativegenetics Cropenergy Biodiesel
a
b
s
t
r
a
c
t
TheaimofthisstudywastoestimatetherepeatabilitycoefficientofgrainproductioninJatropha,the minimumnumberofmeasurementsneededtoreliablypredictthegeneticvalueofselectedfamilies,and todeterminethecumulativegeneticgainswhenconsideringtheselectionofthebestfamiliesbasedon differentnumberofmeasurements.Theexperimentwasconductedwith175accessions(half-siblings progeniesderivedfromselectedplantsinthefield)thatcomposepartofagermplasmcollection.Such bankwasestablishedinarandomizedblockdesignwithtwoblocks.Ineachblockagivenaccessionwas representedina5plant/plotscheme(half-siblings).Fortheanalysis,yielddataobtainedintheyears of2009–2012wereconsidered.Theresultsofthisstudyindicatethattherepeatabilitycoefficientof grainproductioninJatrophaislow(0.37),butcomparabletootherperennialspecies,andthattoachieve reliabilitiesof70and80%inthepredictionofbreedingvaluesofselectedfamilies,4and7yearsof evaluation,respectively,areneeded.Theresultsofthisstudyalsoindicatethattheefficiencyofearly selectionseemstobesmallinJatrophasincethecoincidencerateofselectedgenotypesatearlyage(1or 2yearsofevaluation)andgenotypesselectedinadultage(4yearsassessment)issmall(17–23%).Finally, takingintoaccounttherepeatabilitycoefficientsandcoefficientsofdetermination,inahypothetical periodof21years(whichisequivalenttothreeselectioncyclesusingsevenconsecutivemeasurements –R2=80%),thispaperdemonstratesthathighercumulativegeneticgainscanbeobtained(159%over 108%)byrelaxingthedegreeofcertaintyinpredictingthebestfamilies(R2=65%insteadofR2=80%), sinceitmakespossibletoperformagreaternumberofselectioncyclesinthesameperiod(7cycles insteadof4).
© 2013 Elsevier B.V. All rights reserved.
1. Introduction
Physicnut(JatrophacurcasL.)isanon-edible,subtropical multi-purposecropthatproducesoilbearingseeds,whichcanbeused forawidevarietyofbio-basedmaterialsincludingbiodiesel,biojet fueland specialtychemicals.Becauseofitsnumerouseconomic and sustainable attributes, it has attractedthe interests of the researchandenergysectors(Durãesetal.,2011;Diasetal.,2012
∗ Correspondingauthor.
E-mailaddresses:[email protected](B.G.Laviola),
[email protected](A.M.C.e.Oliveira),[email protected] (L.L.Bhering),[email protected](A.A.Alves),
[email protected](R.B.Rocha),[email protected](B.E.L.Gomes), [email protected](C.D.Cruz).
#4919).Severalgovernmentalincentiveagenciesworldwidehave madeagrowingvolumeofresourcesavailableforresearchand developmentofthespecies(Diasetal.,2012).Moreover,several corporations,fromtheenergysector,arenowpromotingJatropha asoneofthemostviablefeedstocksforlarge-scaleproductionof sustainableplantoilandasoneofthemostpromisingoilplant speciesforbiodieselandbio-kerosene(jet-fuel)production.As con-sequenceoftheseefforts,largecorporationshaveinvestedagreat deal inits wide scale planting(Diaset al.,2012).For instance, ChinaandIndiasolely,havealreadymorethan2.5millionhectares plantedwithJatropha(Fairless,2007),despitethefactthatmostof thegeneticvariabilityofthespeciesisconcentratedinthe Cen-traland South Americas(i.e. Mexico,Colombia,Guatemala and Brazil).IntheseregionsJatrophahasalsoattractedattention.In Brazilforinstance,Jatrophahasbeenconsideredforsomeyears now(Laviolaetal.,2010b)asamajoralternativespeciesthatcan 0926-6690/$–seefrontmatter © 2013 Elsevier B.V. All rights reserved.
complementsoybeansasasourceofvegetableoilforproduction ofbiodieselandbio-kerosene.Inadditiontogoodyield(Drumond etal.,2010)andoilqualityfavorabletotheproductionof biofu-els(Freitasetal.,2011),itswideadaptabilitytodifferentregionsof Brazilanditslongevityhasattractedtheinterestofvariousresearch groups(Durãesetal.,2011).Currentlythereareatleast20,000ha plantedwithJatrophainthecountry,andwiththeinterestofthe privatesectorsthisareaisexpectedtorapidlyincrease.
However,despiteitspotential,thespeciesisstillunder domesti-cation,andtherearenocultivars,norvalidatedproductionsystems fordifferentproducingregionsworldwide.Therefore,researchand developmentinitiativesledbothbythepublicandprivatesectors havefocusedmainlyonbreeding(Durãesetal.,2011;Laviolaand Alves,2011).Severalstudieshavebeenconductedtodetermine physicnutgeneticvariabilityandbreedingpotential(Abdelgadir etal.,2012; Bheringet al.,2012a,b;Gurgeletal., 2011;Laviola etal.,2011,2012c;Mastanetal.,2012;Pandeyetal.,2012;Rocha etal.,2012b;Rosadoetal.,2010;Silva-Junioretal.,2011).Most ofthesestudiesindicatedthat therearegood prospectsforthe species’breedingingiventheopportunitytoselectsuperior materi-als,althoughthegeneticbasisofmaterialontheActiveGermplasm BankofEmbrapaissmall(Rosadoetal.,2010).However,littleis knownabouttherepeatabilityoftargettraits(e.g.grain produc-tioninagesover3years,sincestudiesonthesubjectconsideronly twoyears–Laviolaetal.(2012c).Inbreedingofperennialplants, notonlythedeterminationofvariability,butalsothe repeatabil-ityestimatesoftargettraitsareimportantfordesigningabreeding program,sincetherepeatabilitycoefficientmeasurestheability oforganismsin repeatingthecharacterexpressionover several periodsoftime.Thisallowsverifyingwhetherthesuperiorityof somegenotypesismaintainedovertheyears,orwhetheritwasdue tosometransientenvironmentalcondition.Highvaluesof repeat-abilityindicatethatitispossibletopredicttheactualbreedingvalue oftheindividualbasedonfewsequentialmeasurements.Froma practicalstandpoint,thisparameterpresentscrucialimportancein predictinggeneticandgenotypicvaluesandintheinferenceabout theincreaseofselectiveefficiencybyusinganestablishednumber ofmeasurementsperindividual,whichallowthedeterminationof thenumberofcropstobeadoptedinabreedingprogram(Resende, 2002).
Therepeatabilitycanvarydependingonthenatureofthetrait, onthegeneticpropertiesofthepopulation,andonthe environ-mentalconditionsunderwhichindividualsaremaintained(Cruz etal., 2004).Moreover,thereare differentmethods for obtain-ingrepeatabilitycoefficientestimates.Thesemethodshavebeen usedinseveralperennialspecies,withdifferentapplicationsand particularities.Themethodofanalysisofvariance,forexample,is indicatedinordertoevaluatepgenotypesinnrepeatedmeasures, consideringthepropertiesandconstraintsofanevaluationusing theleast-squaresmethod(Cruzetal.,2004).Ontheotherhand, methodsbasedonprincipalcomponentsareconsideredthemost appropriateinsituationswhentheevaluatedgenotypespresent cyclicalbehaviorinrelationtothetraitstudied,andwhentheydo notmeettheassumptionsofvariancehomogeneityandrandom distributionofresidues(Abeywardena,1972;Rutledge,1974).On theotherhand,themethodofstructuralanalysis(Mansour,1981) isalsodifferentiatedsinceitrequiresfewassumptions,anddiffers itselffromthemethodofprincipalcomponentsonlybyconceptual issues(Cruzetal.,2004).
Basedontheaforesaid,theaimofthisstudywas:(i)to deter-minetherepeatabilitycoefficientofgrainproductioninJatropha (consideringevaluationsoverfourconsecutiveyears);(ii)to estab-lish,basedonthecoefficientsofrepeatabilityanddetermination, theminimumnumber ofmeasurementsneededtopredictwith predeterminedreliabilitythegeneticvalueoftheselected fam-ilies;(iii) to verifythe coincidence in theselection of thebest
families carriedout indifferentyears;and finally(iv)to deter-minethecumulativegeneticgainwhenconsideringtheselection ofthebestfamiliesbasedondifferentmeasurementnumbers(i.e. withdifferentdegreesofcertaintyinpredictingthebestfamilies). Theseresults,togetherwiththosepreviouslypublished,canhelp toestablishstrategiesforJatrophabreedingthatenablestherapid developmentofmoreproductivevarieties/cultivars.
2. Materialsandmethods
2.1. Plantmaterial,experimentaldesignandevaluation
Theexperimentwasconductedwith175Jatrophaaccessions (half-siblingsprogeniesderivedfromselectedplantsinthefield) from the germplasm bank of Embrapa Agroenergy, which is installedintheexperimentalareaofEmbrapaCerrados,Planaltina, DF,Brazil(lat.15◦3530S,long.47◦4230W,and1007malt.asl). Theregionpresentsatropicalclimatewithdrywinterandrainy summer.Theaveragetemperatureis22◦Candthemeanrelative humidityis73%.Thetotalannualrainfallisabout1000mm.The predominantsoilinthelocationwasclassifiedasOxisolwithhigh claycontent.Thegermplasmbankwasestablishedin2008ina randomizedblockdesignwithtwoblocks.Ineachblockagiven accessionisrepresentedina5plant/plotscheme(half-siblings). Theplotswerearrangedinrows,spaced4mapart,andeachplant wasplaced2mapartfromthenextplantintherow.Inorderto determinetherepeatabilitycoefficient,thedataobtainedfordry grainsproduction,infouragriculturalyearswereconsidered.Data wascollectedbetweenJanuarytoJune of2009,2010,2011and 2012.Theplantmanagementwasperformedaccordingtothe lat-estresearchresults(AnithaandVaraprasad,2012;Karanamand Bhavanasi,2012;Laviolaetal.,2012a;Resendeetal.,2012a,b). 2.2. Estimatesofrepeatabilitycoefficients
Data weresubjected toanalysis ofvariance andthe genetic parameters estimated consideringhalf-sib families, information within theplotand specific environmentalconditions in Brazil (cerradoconditions–tropicalclimatewithdrywinterandrainy summer).Toobtaintheestimateoftherepeatabilitycoefficient, datawerepreviouslyclassifiedwithineachmeasurement, accord-ingtoCruz(2006b).Thecalculationoftherepeatabilitycoefficient wasthenperformedbythefollowingmethods:(i)analysisof vari-ance, in which therepeatability coefficientis estimated by the resultsof analysisofvariance; (ii)principalcomponents,based on thecovariance matrix byapplying the matrices of variance andphenotypiccovariances;(iii)principalcomponents,basedon thecorrelationmatrix,inwhichtheestimatorofrepeatabilityis obtainedbasedontheassumptionthattherepeatabilitycoefficient isgiven bythecorrelationbetweeneachpairofmeasurements evaluatedindifferentgenotypes;and(iv)thestructuralanalysis basedoncorrelationmatrix,inwhichtheparametriccorrelation matrixbetweenthegenotypesisconsidered,ineachpairof eval-uation,withrepeatabilitycoefficientestimatorbasedonstructural analysis.Fordetailsaboutthemethods,seeCruz(2006a). 2.3. Selectiongainandcoincidenceofselectedfamilies
Inadditiontotheestimationofrepeatability,aselection inten-sity of 10% was applied aiming to estimate the coefficient of coincidenceintheselectionofthe17bestgenotypes(mass selec-tion)betweenyears of evaluation.A dispersiongraph wasalso generatedforgenotypesbehaviorbetweenthefirsttwoyears(P12)
andinthelasttwoyearsofevaluation(P34)inordertovisualize
theindividualandoverallperformanceofthegenotypesoverthe firstfouryears(andtoidentifyfamilieswithincreasedphenotypic
Table1
Analysisofvarianceofgrainyield(gplant−1)generatedaftertheevaluationof175
Jatrophaaccessionsduringfourconsecutiveagriculturalyears,demonstratingthe existenceofgeneticvariabilitythatcanbeexploitedbythebreedingprograms.
SV DF MS Blocks 1 1,599,740.50 Genotypes(G) 174 56,548.69* Years(Y) 3 49,038,970.32* G×Y 522 32,718.10* Residue 696 18,381.64 Total 1399
SV,sourceofvariation;DF,degreesoffreedom;MS,meansquare.
* Significantat1%probability.
stabilityoverthefouryears).Allanalyseswereperformedusing theGENESsoftware(Cruz,2006a).
3. Resultsanddiscussion
3.1. Variability,heritabilityandcoefficientofexperimental variationforgrainproduction
Thefactthat thisworkwasbasedona four-yearevaluation (2009–2012)makesitoneofthemostcompletestudiesevermade inthecultureofJatrophatodate,eventhoughitaddressspecific environmentalconditionsinBrazil.Inthiscontext,itisimportant mentioningthattheresultsobtainedbyANOVAindicatethe exist-enceofgeneticvariabilitythatcanbeexploitedbythebreeding program,assuminga1%levelofprobabilityforgrainproduction betweenthe175genotypesevaluated(Table1).Thisisextremely relevant,sinceitconfirmspreviousstudiescarriedoutwhilethe plantswereatearlyages.Suchstudies(Bheringetal.,2012b,2013; Laviolaetal.,2010a,b,2012b,c)hadalreadydemonstratedthe exist-enceofgeneticvariabilityforgrainproduction,andtheprospectof selectiongainforgrainproductioninthesamepopulation,even withalimitedgeneticbase(Rosadoetal.,2010).Itcanbeargued thatnotonlygrainproductionisanimportanttraitinthiscrop, andthatothertraitsshouldbeevaluated,butgiventhefactthat improvedcultivarshavenotbeendevelopedyet,thistraitcertainly isthemostimportantinevaluations.Thisiscorroboratedbythe factthatresearchresultspublishedsofar,indicatetheexistence ofalooseconnectionbetweentheyieldcomponentswithoil pro-duction(Bheringetal.,2012b,2013;Laviolaetal.,2010b,2011, 2012b,c;Rochaetal.,2012a;Spinellietal.,2010).Thus,oneshould focusdirectlyonbreedingforgrainproductiontoincreasetheoil productivityperhectare.
Moreover,withrespecttograinyield, it canbeobserved in Table2thatingeneralthereisanupwardtrendinthepopulation’s averageovertheyears,mainlybetweenthethirdandfourth har-vest.Thistrendisalsoaccompaniedbyadownwardtrendinthe experimentalcoefficientofvariationovertheyears.Suchtrends mustberelated(i)tothefactthatintheearlystagesofgrowth, plantsaremoresensitivetoenvironmentalvariationsand(ii)to thefactthatyoungperennialplantsoftenhavetheirmetabolism towardvegetativeratherthanreproductivegrowth(Larcher,2004).
Table2
Mean(gplant−1),broad-senseheritability(h2)andcoefficientofvariation(CV%)for
grainyieldof175Jatrophaaccessionsevaluatedduringfourconsecutiveagricultural years.
Agriculturalyear(plantsage) Mean(gplant−1) h2 CV(%)
2009(1yr) 11.32 0.69 76.43 2010(2yr) 175.94 0.62 32.23 2011(3yr) 329.73 0.31 33.57 2012(4yr) 874.35 0.54 27.54
Table3
Repeatabilitycoefficients(r)andcoefficientsofdetermination(R2)forgrainyield
obtainedbydifferentstatisticalmethods,aftertheevaluationof175Jatropha acces-sionsevaluatedduringfourconsecutiveagriculturalyears.
Method r R2(%)
ANOVA/structural–COV 0.37 70.18 Principalcomponents–COV 0.37 70.25 Principalcomponents–COR 0.37 70.25
Structural–COR 0.37 70.18
COV,methodofcovariance;COR,methodofcorrelations.
Theheritability,initsbroad-sense,alsoshowedadownwardtrend over theyears. However,it mustbeconsideredthat the coeffi-cientofheritabilityestimatedonyoungperennialplantsareoften inflatedbygenotype×yearinteractions(Resendeetal.,2001).In fact,whenanalyzingtheeffectofgenotype×yearinteractionsit is noticedthat it wasalsosignificantat 1% levelof probability (Table1),indicatingtheexistenceofgenotypeswithdifferent per-formanceovertheyearsintermsofgrainproduction.Thisfactwas expected,sincethesignificantinteractionbetweengenotypesand yearsisrecurrentinevaluationofperennialplants,anditis com-monlycausedbytheeffectoftheenvironmentontraitexpression (Resende,2002).Thus,theseresultsindicatethatitisnecessary thatevaluationsarecarriedoutovermanyyearsuntilthegrain productionisstable,sinceJatrophaisaperennialspecies.Inthis context,thisisoneofthefirststudiestobringtheresultsofcareful evaluationofJatrophaplantsolderthan36months.
3.2. Repeatabilitycoefficientandminimumnumberof measurements
Fromthedatacollectedinfouryearsofevaluation(2009–2012) and based on different biometric methods, the coefficients of repeatability and determination were determined (Table 3).As mentionedbefore,therepeatabilitycoefficientmeasurestheability oforganismstorepeattheexpressionofatraitoverseveralperiods oftime,whereasthecoefficientofdetermination(R2)measures
thedegreeofcertaintyinpredictingtherealvalueofanindividual. Itwasobservedthat,regardlessofthemethod,therepeatability coefficient(r)didnotvary,indicatingthattheadoptionofoneor anotherstrategydoesnotaffecttheestimate.Thisresultis interest-ing,sinceinsomecasestheoptionforeithermethodcanimprove theefficiencyof breedingprograms.Forexample,Bheringetal. (2013),whencomparingdifferentselectionmethodsinJatropha, verifiedthat the combinedselection is more suited than other methodsforrapidimprovementofthisspecies.
Inrelationtotheestimatedvalueoftherepeatabilitycoefficient, thisvalueisconsideredlow,especiallywhencomparingit with thevaluesfoundforthissamepopulation(whenonlytwo evalua-tionswereconsidered)(Laviolaetal.,2012c).Nevertheless,when comparingthevalueobtainedwiththerepeatabilitycoefficient forproductioninotherperennialspecies,itisobservedthatitisin factsomewhatsimilar.Thisisbecausethegenotypestabilization didnotoccuruntilthefourthharvestyear.AccordingtoCruzetal. (2004),grainproductionisacomplextraitanditisdeterminedby differentgenepools,anditispossiblethatdifferentgenepoolsare expressedatdifferentstages,andevenwithinagivengroup,some genesmaybemoreorlessexpressed,accordingtothe develop-mentalstageofthegenotypes.Thus,whenarepeatabilitystudyis conductedingenotypeswhicharenotstabilizedyet,low repeat-abilitymaybefound,whichdoesnotmeanthatthesolutionforthe problemistheincreaseinthenumberofrepetitions.Insomecases, theabsenceofevaluationinearlystages,inwhichthereisnofull manifestationofthegeneticpotentialofthematerialstudied,may increasetheestimateofrepeatability.Inthisstudy,asitisshown inTable2,thecoefficientofenvironmentalvariationinthefirst
Fig.1. Graphicanalysisoftheminimumnumberofmeasurementsrequiredtoreach adetermineddegreeofcertaintyintheselectionofthebestgenotypesofJatropha forgrainproductionaccordingtodifferentmethods(analysisofvariance,principal componentsbasedonthecovariancematrix,principalcomponentsbasedonthe correlationmatrixandstructuralanalysis).
seasonwasquiteevident,whichmayhavecontributedtoreduce therepeatabilitycoefficientofgrainproduction.Inrelationtothe studyof Laviola etal. (2012c),in which highrepeatabilitywas verified,thelowerrangeofvariationofthemeanandCVmayhave contributedtotheresult.Thus,infuturestudies,theexclusionof thefirstseasonandtheothersbelongingtoperiodsinwhichthereis nogenotypicstabilizationmaybeamoreappropriatewayfor esti-matingthecoefficientofrepeatabilityandpredictingwithgreater accuracythegeneticvalueofJatrophafamiliesandindividuals.
Inrelationtotheestimatedcoefficientofdetermination,itis observedthatexceptforminordecimalvariations,theresultwas essentiallythesame,regardlessofthemethodologyused(Table3). Itisnoteworthythatthevaluefoundallowsgoodreliabilityinthe predictionandselectionofgenotypes.Theevolutionofthe mini-mumnumberofmeasurementsrelatedtothedegreeofcertainty (R2)intheselectionofthebestgenotypes(predictionoftheactual
value)isshowninFig.1.Accordingtothisresult,theuseoffour repeatedmeasurements leadsto a coefficient of determination of70%,whereassevensequentialmeasurementsarerequiredto achieveanaccuracyof80%.Theseresultsareconsistentwiththose publishedforotherperennialspeciessuchasoilpalmand euca-lypts(Chiaetal.,2009)andalsowiththosepublishedbyLaviola etal.(2012c),whichshowedthatatleastfourmeasurementswere necessaryforachievingreasonableselectiveaccuraciesbasedon theevaluationofyoungplants.
3.3. Coincidenceintheselectionofthebestfamiliesindifferent years
Asthereisthepossibilityofselectingthebestfamiliesbased ondifferentnumbersofevaluations,itisimportanttoknowifthe breederwillbeselectingthesamegeneticmaterialwhiledoing theselectionbasedondifferentstrategies.Inordertoexplainthis issue,thetop17genotypes(selectedconsideringaselection inten-sityof10%)selectedforeachyearwerecompared(Table4)using thecoefficientofcoincidence.Inordertocomputethecoefficient ofcoincidenceintheselectionofthe17bestgenotypes,different combinationswereconsidered:year×year,year×selectionbased ontheaveragesoffouryears(M),andselectionbasedonthemean ofthefirsttwoyears(P12)×selectionbasedonthemeanofthelast
twoyears(P34)(Table5).
Ingeneral,plantswhichstandoutinoneoftheyearsarenot thesameplantsthatpresent bestperformance inthefollowing
Table4
SelectedsetsofJatrophagenotypesbasedongrainyield(gplant−1)consideringthe
informationofeachyearindividually,thefouryearsaverage(P1234)andthemean
ofthefirsttwoyears(P12)andofthelasttwoyears(P34). Years SelectedJatrophaindividuals
1 6174751264633321012214310212323173103 2 61741733117210183351402562761583347123165 3 4638151717617289348212221947 4 31722331597310216325155941391541111578129 P1234 31722338115949425611557310213930512 P12 6174173311721018335140336281257612315847 P34 31722433159817394155652510215413930
P1234,meanoffouryears;P12,meanofthefirsttwoyears;P34,meanofthelasttwo
yearsofevaluation.
year(Table4).Thisbehaviortendstolimittheselectionofplants in thejuvenile phase,based ontheevaluation of oneor a few years,sincetheplantwhichhasbetteryieldpotentialistheone thatmaintainsitssuperiorperformanceafterstabilizingits pro-ductivity(Cavalcanteetal.,2012;Cruzetal.,2004;Resende,2002). Thisobservationiscorroboratedbytheestimatedcoefficientof coincidence.Thecoincidenceofselectedgenotypesislowin all combinationsyear×year,indicatingthat thereisalow mainte-nanceofproductivestabilityofgenotypesinthefirstfouryears (Table5).
Themagnitudeofthecoefficientofcoincidencebetween geno-typesselectedinthefirstandlastyearpermitstheevaluationofthe efficiencyofearlyselection.Inyear×yearcomparisons,itwouldbe interestingtofindhighcoefficientofcoincidenceamonggenotypes selectedinearlyages(1or2yearsofevaluation)andthoseselected atthelastyearofevaluation(representingadulthood),asthismight allowearlyselectionofsuperiorfamilies/individuals.However,the coincidenceofthesecomparisonswaslow(Table5),which demon-stratesthattheselectionbasedonasingleseasonandatagesequal orlessthan3yearsdoesnotreflecttheproductionatolderages. Theseresultsconfirmthelowrepeatabilitycoefficientfoundinthe 4initialseasons(Table3).Injuvenilephase,perennialplantsmay presentgreatvariationamonggenotypesfor(re)production,since inthisperiodthemajorityofexpressedgenesareassociatedwith theformationofvegetative organsandpresent greaterstability atolderages(Larcher,2004).Thecomparisonamong the geno-typesselectedinthe4thyear(A4)andthegenotypesselectedusing informationfromallyears(Table5)presentedthehighest coeffi-cientofcoincidence(70.59%).Thisresultistypicalofperennials, whoseannualgrainproductionisstillonanupwardtrend,being thelastharvestthemostimportantin theselection ofthebest families.
Table5
Coefficientofcoincidenceintheselectionforthe17bestaccessionsforgrainyield (gplant−1)consideringthecombinationoftwoconsecutiveyears(A×A),the com-binationofindividualyearswiththemeanofthefourconsecutiveyears(A×M)and theaverageofthefirsttwowiththeaverageofthelasttwoyears(P12×P34).
Measurements Coincidence(n) Coincidence(%)
A1×A2 6 35.29 A1×A3 2 11.76 A1×A4 4 23.53 A2×A3 2 11.76 A2×A4 3 17.64 A3×A4 3 17.64 A1×M 6 35.29 A2×M 4 23.53 A3×M 5 29.41 A4×M 12 70.59 P12×P34 4 23.53
An,year;M,meanoffouryears;P12,meanofthefirsttwoyearsofevaluation;P34,
Fig.2. Dispersionoftheperformanceofthe175Jatrophagenotypesoverthefirstandthelasttwoyearsofcultivation.Xaxis–(P12)andyaxis–(P34).
Thedispersionoftheperformanceratioofthe175genotypesin thefirsttwoyears(P12)andinthelasttwoyears(P34)isalsoan interestingevaluationforidentifyingindividualandoverall per-formanceof genotypesover thefirst fouryears (Fig. 2).In this graph,genotypesareplotted,onthexaxis,accordingto produc-tion(gplant−1)inthefirsttwoyears;andaccordingtoproduction (gplant−1)inthelasttwoyears,ontheyaxis.Thus,inquadranttwo (QII),genotypesthatshowedabove-averageperformanceinthe firstandlasttwoyearsarepresented.Therefore,familiesobserved inthisquadrantarethosethatpresentedgreaterphenotypic sta-bilityoverthefouryears.Examplesaregenotypes33,81and25, withaverageyield,ingplant−1,of159.38,155.48and154.98(first andsecondyear);and905.60,885.06and835.36(thirdandfourth year),respectively.InquadrantI(QI),thegenotypesthatshowed lowaverageyieldinthefirstandsecondyearareshown; never-theless,theystoodoutfortheirhighaverageyieldinthethirdand fourthyear.Genotype7ishighlightedduetoitslowyieldinthefirst twoyears,butitwasthegeneticmaterialthatpresentedthehighest averageinthisquadrant.Yieldsshownduringfourthyeararethose thatcontributemosttotheselectionofsuperiorgenotypes.Thus, theselectionofgenotypesinQIbecomesquiteinteresting,since eventhoughthegenotypespresentedbelow-averageperformance inthefirsttwoyearstheypresentedhighestyields.Finally,in quad-rantIIIandIV,thelowyieldfamiliesarepresented(below-average production),intheevaluationsoftheyieldintheyearsP12andP34.
Alargenumberofmaterialscomposethesequadrantsandahigh concentrationandoverlapofgenotypesisobservedinquadrantIII. Thesegenotypesshouldbediscardedintheprocessofbreedingfor grainproduction,sincetheycontributelittletothedevelopmentof productivecultivars.
3.4. Cumulativegeneticgainconsideringdifferentstrategies accordingtothenumberofmeasurements
Providedthat3additionalmeasurementswouldbenecessary toachieve80%accuracy,besidesthefourmeasurementsnecessary
for70%accuracy(Table6),itisimportanttoassesswhethera10% increaseinselectionefficiencyjustifiesanincreaseof75%ofthe timetoconcludetheselection cycle(7vs.4years). Inorderto makethiscomparison,thegeneticgainwascalculatedbyapplying pressureof10%.Suchgainwaslateradjustedconsideringdifferent selectionefficiency(whichcorrespondstoR2 –65,70and80%).
Tomake this comparisonmore realistic,it wasconsideredthat withthe narrowing of thegenetic baseafter a selection cycle, theselectiongain in advancedgenerations would bedecreased by5%.Table6showsaprojectionofexpectedgeneticgainsina Jatrophabreedingprogramafter21yearsofbreeding.Table6also showsthat,dependingontheaccuracyadoptedbythebreeder, withinapredeterminedperiodof21years,from3upto7selection cyclesforgrainproductioncanbeperformed.Furthermore,higher cumulative geneticgains canbe obtainedbyrelaxing thelevel ofcertaintyinthepredictionofbestfamilies,sincethetotalgain
Table6
SimulationindicatingthegeneticgainsexpectedforJatrophagrainyield(gplant−1)
insuccessiveselectivecycles,thenumberofselectioncyclespossibletobe per-formed in apredetermined hypotheticalperiodof 21years of breedingand accumulatedgeneticgainforJatrophainrelationtotheadoptionofdifferent coefficientsofdeterminationbythebreeder.
Selectivecycle Expectedgain(%) Adjustedgaina
R2=65%b R2=70%c R2=80%d 1 50 32.5 35.0 40.0 2 45 29.2 31.5 36.0 3 40 26.0 28.0 32.0 4 35 22.7 24.5 5 30 19.5 21.0 6 25 16.2 7 20 13.0 Accumulatedgain(%) 159.2 140.0 108.0
aAdjustedgainR2=Gain×R2. b3yearsofevaluation/cycle. c 4yearsofevaluation/cycle. d 7yearsofevaluation/cycle.
providedbythestrategyusing65%ofcertaintyis47%higherthan thatprovidedbyadoptingastrategywith80%ofaccuracy.Through thesecomparisons,theadoptionofthreeorfourmeasurements in time(yield evaluations),despite havinglowercoefficients of determination, may allow greater accumulated genetic gains overseveralselectioncycles. It canbearguedthat carryingout 7selectioncyclesin21years,insteadofjustthree,mayleadto rapiddepletionofgeneticvariability.Thisis arelevant concern indeed.However,aswelldiscussedbyBheringetal.(2013)(when comparingtheexpectedgainsusingdifferentselectionstrategies), consideringthe growingdemandfor improved varietiesof Jat-rophainBrazilandabroad,thereisa trendtothefirststrategy, sinceJatrophabreedingprogramshavetorespondquicklytothis demand by developing more productive varieties. Considering thatthegeneticbasisofJatrophainBrazilisalreadyconsidered tobelimited(Rosadoetal.,2010),theadditionofnewsourcesof variability willbeimperative in short/mediumterm, regardless of7or3selectioncycles,sincebreedersshouldnotwaituntilthe geneticbasisisexhaustedtothenseekforasolution.
4. Conclusions
Ingeneral,theresultsofthisstudyallowustoconcludethat: (i)therepeatabilitycoefficientofgrainproductioninJatrophais low,butit iscomparabletootherperennialspecies,suchasoil palmandeucalypts,(ii)basedontherepeatabilitycoefficient,4and 7aretheminimumnumberofmeasurementsrequiredtopredict thegeneticvalueoftheselectedfamilieswithreliabilitiesof70and 80%,respectively,intheevaluatedenvironmentalconditions,(iii) thecoincidenceintheselectionofthebestfamiliescarriedoutin differentyearsislow;therefore,theefficiencyofearlyselection issmall,and(v)greatercumulativegeneticgainscanbeobtained byrelaxingthedegreeofcertaintyinthepredictionofthebest families.
Acknowledgements
The authors acknowledge FINEP for supporting the project “Research,DevelopmentandInnovationinJatrophaforBiodiesel Production (BRJATROPHA)”, and CNPq, FUNARBE, CAPES and FAPEMIGforprovidingscholarshipsforstudents,andforresearch support.TheauthorsthankthestaffofEmbrapaAgroenergyJulio CesarMarana,LaiseTeixeiradaCostaandGenivaldoJoseFonseca forthededication inconductingofmanagement andevaluation workofthegermplasmbank.
References
Abdelgadir,H.A.,Johnson,S.D.,VanStaden,J.,2012.Pollenviability,pollen ger-minationandpollentubegrowth inthebiofuelseedcrop Jatrophacurcas (Euphorbiaceae).S.Afr.J.Bot.79,132–139.
Abeywardena,V.,1972.Anapplicationofprincipalcomponentanalysisingenetics. J.Genet.61,27–51.
Anitha,K.,Varaprasad,K.S.,2012.PhytosanitaryaspectsofJatrophaframinginBrazil. In:Carels,N.,Sujatha,M.,Bahadur,B.(Eds.),Jatropha,ChallengesForaNew EnergyCrop.Springer,NewYork,pp.175–218.
Bhering,L.L.,Barrera,C.F.,Ortega,D.,Laviola,B.G.,Alves,A.A.,Rosado,T.B.,Cruz, C.D.,2012a.DifferentialresponseofJatrophagenotypestodifferentselection methodsindicatesthatcombinedselectionismoresuitedthanothermethods forrapidimprovementofthespecies.Ind.CropProd.41,260–265.
Bhering,L.L.,Barrera,C.F.,Ortega,D.,Laviola,B.G.,Alves,A.A.,Rosado,T.B.,Cruz,C.D., 2013.DifferentialresponseofJatrophagenotypestodifferentselectionmethods indicatesthatcombinedselectionismoresuitedthanothermethodsforrapid improvementofthespecies.Ind.CropProd.41,260–265.
Bhering,L.L.,Laviola,B.G.,Salgado,C.C.,Sanchez,C.F.B.,Rosado,T.B.,Alves,A.A., 2012b.Geneticgainsinphysicnutusingselectionindexes.Pesq.Agropec.Bras. 47,402–408.
Cavalcante,M.,Lira,M.A.,Santos,M.V.F.,Pita,E.B.A.F.,Ferreira,R.L.C.,Tabosa,J.N., 2012.Coeficientederepetibilidadeeparâmetrosgenéticosemcapim-elefante. Pesq.Agropec.Bras.47,569–575.
Chia,G.S.,Lopes,R.,Cunha,R.N.V.,Rocha,R.N.C.,Lopes,M.T.G.,2009.Repeatability forbunchproductionininterspecifichybridsbetweenCaiauéandAfricanoil palm.ActaAmazonica39,249–254.
Cruz,C.D.,2006a.ProgramaGenes:Biometria.EditoraUFV,Vic¸osa,MG.
Cruz,C.D.,2006b.ProgramaGenes:EstatísticaExperimental.EditoraUFV,Vic¸osa, MG.
Cruz,C.D.,Regazzi,A.J.,Carneiro,P.C.S.,2004.Modelosbiométricosaplicadoao mel-horamentogenético,3rded.UFV,Vic¸osa.
Dias, L.A.S., Missio, R.F., Dias, D.C.F.D., 2012. Antiquity, botany, ori-gin and domestication of Jatropha curcas (Euphorbiaceae), a plant species with potential for biodiesel production. Genet. Mol. Res. 11, 2719–2728.
Drumond, M.A., Santos, C.A.F.,de Oliveira, V.R.,Martins, J.C., dos Anjos, J.B., Evangelista,M.R.V.,2010.Agronomicperformanceofdifferentgenotypesof physic nut in the semi-arid zone of Pernambuco state. Cienc. Rural 40, 44–47.
Durães,F.O.M.,Laviola,B.G.,Alves,A.A.,2011.Potentialandchallengesinmaking physicnut(JatrophacurcasL.)aviablebiofuelcrop:theBrazilianperspective. Persp.Agric.Vet.Sci.Nutr.Nat.Resour.6,8.
Fairless,D.,2007.Biofuel:thelittleshrubthatcould–maybe.Nat.Biotechnol.449, 652–655.
Freitas,R.G.,Missio,R.F.,Matos,F.S.,Resende,M.D.V.,Dias,L.A.S.,2011.Genetic eval-uationofJatrophacurcas:animportantoilseedforbiodieselproduction.Genet. Mol.Res.10,1490–1498.
Gurgel,F.D.,Laviola,B.G.,daSilva,D.M.,Kobayashi,A.K.,Rosado,T.B.,2011. Pheno-logyandbiometryofphysicnutintheBrazilianSavannah.Cienc.Agrotec.35, 1122–1131.
Karanam,K.R.,Bhavanasi,J.K.,2012.CultivationtechnologyforJatrophacurcas.In: Carels,N.,Sujatha,M.,Bahadur,B.(Eds.),Jatropha,ChallengesforaNewEnergy Crop.Springer,NewYork,pp.165–174.
Larcher,W.,2004.EcofisiologiaVegetal.Rima,SãoCarlos.
Laviola, B.G., Alonso, A.A., Rocha, r.b., Drumond, M.A., 2012a. The impor-tance of Jatropha for Brazil. In: Carels, N., Sujatha, B., Bahadur, B. (Eds.), Jatropha, Challenges for a New EnergyCrop. Springer, New York, pp.71–94.
Laviola,B.G.,Alves,A.A.,2011.Matérias-primasoleaginosasparabiorrefinarias. In:Vaz,J.S.(Ed.),Biorrefinarias:cenárioseperpectivas.EmbrapaAgroenergia, Brasília,pp.29–43.
Laviola,B.G.,Alves,A.A.,Gurgel,F.D.,Rosado,T.B.,Costa,R.D.,Rocha,R.B.,2012b. Estimateofgeneticparametersandpredictedgainswithearlyselectionofphysic nutfamilies.Cienc.Agrotec.36,163–170.
Laviola,B.G.,Alves,A.A.,Gurgel,F.D.,Rosado,T.B.,Rocha,R.B.,Albrecht,J.C.,2012c. Estimatesofgeneticparametersforphysicnuttraitsbasedinthegermplasm twoyearsevaluation.Cienc.Rural42,429–435.
Laviola,B.G.,Bhering,L.L.,Mendonca,S.,Rosado,T.B.,Albrecht,J.C.,2011. Morpho-agronomiccharacterizationofthegermplasmbankofJatrophayoungstage. Biosci.J.27,371–379.
Laviola, B.G., Rosado, T.B., Bhering, L.L., Kobayashi, A.K., de Resende, M.D.V., 2010a. Genetic parameters and variability in physic nut acces-sions during early developmental stages. Pesq. Agropec. Bras. 45, 1117–1123.
Laviola, B.G., Rosado, T.B., Bhering, L.L., Kobayashi, A.K., Resende, M.D.V., 2010b. Genetic parameters and variability in physic nut acces-sions during early developmental stages. Pesq. Agropec. Bras. 45, 1117–1123.
Mansour, H., 1981. Estimator of repeatability. Theor. Appl. Genet. 60, 151–156.
Mastan,S.G.,Sudheer,P.D.V.N.,Rahman,H.,Reddy,M.P.,Chikara,J.,2012. Develop-mentofSCARmarkerspecifictonon-toxicJatrophacurcasL.anddesigninga novelmultiplexingPCRalongwithnrDNAITSprimerstocircumventthefalse negativedetection.Mol.Biotechnol.50,57–61.
Pandey,V.C.,Singh,K.,Singh,J.S.,Kumar,A.,Singh,B.,Singh,R.P.,2012.Jatropha curcas:apotentialbiofuelplantforsustainableenvironmentaldevelopment. Renew.Sust.Energ.Rev.16,2870–2883.
Resende,J.C.F.,Gonc¸alves,N.P.,Dias,M.S.C.,Albuquerque,C.J.B.,Morais,D.L.B.,2012a. PhytosanitaryaspectsofJatrophafarminginBrazil.In:Carels,N.,Sujatha,M., Bahadur,B.(Eds.),Jatropha,ChallengesforaNewCrop.Springer,NewYork, pp.219–238.
Resende,J.C.F.,Silva,J.T.S.,Simão,F.R.,Pimentel,R.M.A.,Morais,D.L.B.,2012b. Phy-totechnicalaspectsofJatrophafarminginBrazil.In:Carels,N.,Sujatha,m., Bahadur,B.(Eds.),Jatropha,ChallengesforaNewEnergyCrop.Springer,New York,pp.239–262.
Resende,M.D.V.,2002.Genéticabiométricaeestatísticanomelhoramentode plan-tas.Embrapa,Brasília.
Resende, M.D.V., Furlani-Júnior, E., Moraes, M.L.T., Fazuoli, L.C., 2001. Esti-mativas de parâmetros genéticos e predic¸ão de valores genotípicos no melhoramento do cafeeiro pelo procedimento Reml/Blup. Bragantia 60, 185–193.
Rocha,R.B.,Ramalho,A.R.,Teixeira,A.L.,Laviola,B.G.,daSilva,F.C.G.,Militao,J., 2012a.Selectionefficiencyforincreasingphysicnutoilcontent.Pesq.Agropec. Bras.47,44–50.
Rocha,R.B.,Ramalho,A.R.,Teixeira,A.L.,Laviola,B.G.,daSilva,F.C.G.,Militao,J.S.L.T., 2012b.Selectionefficiencyforincreasingphysicnutoilcontent.Pesq.Agropec. Bras.47,44–50.
Rosado,T.B.,Laviola,B.G.,Faria,D.A.,Pappas,M.R.,Bhering,L.L.,Quirino,B., Grat-tapaglia,D.,2010.Molecularmarkersreveallimitedgeneticdiversityinalarge
germplasmcollectionofthebiofuelcropJatrophacurcasL.inBrazil.CropSci.50, 2372–2382.
Rutledge,J.J.,1974.AscalingwhichremovesbiasofAbaywardenas’sestimatorof repeatability.J.Genet.61,247–250.
Silva-Junior,O.,Rosado,T.,Laviola,B.,Pappas,M.,Pappas,G.,Grattapaglia,D., 2011.Genome-wideSNPdiscoveryfromapooledsampleofaccessionsofthe
biofuelplantJatrophacurcasbasedonwhole-transcriptomeIllumina resequen-cing.BMCProc.5,P57.
Spinelli,V.M.,Rocha,R.B.,Ramalho,A.R.,Marcolan,A.L.,VieiraJúnior,J.R., Fer-nandes,C.F.,Militão,J.S.L.T.,Dias,L.A.S.,2010.Primaryandsecondaryyield componentsofthe oilinphysicnut(Jatropha curcasL.). Cienc. Rural40, 1752–1758.
